from sklearn.metrics import confusion_matrix, roc_auc_score, roc_curve, ConfusionMatrixDisplay

def plot_confusion_matrix_and_roc(model,df):

    """Helper function for plotting model performances"""    

    # Make predictions on the holdout set

    X,y = df.drop('target',axis=1), df['target']

    y_pred = model.predict(X)

    # Create the confusion matrix
    cm = confusion_matrix(y, y_pred)

    # Calculate the AUC ROC score
    auc = roc_auc_score(y, y_pred)

    # Get the FPR and TPR for the ROC curve
    fpr, tpr, thresholds = roc_curve(y, y_pred)

    # Create a figure with two subplots
    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 5))

    # Plot the confusion matrix
    ax1.matshow(cm, cmap='Blues')
    ax1.set_title('Confusion Matrix')
    ax1.set_xlabel('Predicted')
    ax1.set_ylabel('Actual')

        # Add labels to the confusion matrix
    # for i in range(cm.shape[0]):
    #     for j in range(cm.shape[1]):
    #         ax1.text(j, i, f'{cm[i, j]:n}', ha='center', va='center')

    # Add labels for the TP, FP, TN, and FN cells
    ax1.text(0, 0, f'TN: {cm[0, 0]:n}', ha='center', va='center', color='w')
    ax1.text(0, 1, f'FN: {cm[1, 0]:n}', ha='center', va='center', color='k')
    ax1.text(1, 0, f'FP: {cm[0, 1]:n}', ha='center', va='center', color='k')
    ax1.text(1, 1, f'TP: {cm[1, 1]:n}', ha='center', va='center', color='k')

    # Plot the ROC curve
    ax2.plot(fpr, tpr, label='AUC ROC = %0.2f' % auc)
    ax2.plot([0, 1], [0, 1], 'k--')
    ax2.set_title('ROC Curve')
    ax2.set_xlabel('FPR')
    ax2.set_ylabel('TPR')
    ax2.legend()

    plt.show()

#train the model

from xgboost import XGBClassifier

model = XGBClassifier()
model.fit(X_resampled,y_resampled)

plot_confusion_matrix_and_roc(model,df_holdout)